Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +9 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-1152.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-1440.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-1728.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2016.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2304.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2592.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-288.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2880.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-3168.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-3456.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-3744.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4032.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4320.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4608.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4896.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-5184.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-5472.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-576.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-5760.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6048.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6336.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6624.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6912.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-72.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-7200.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/checkpoint-864.json +32 -0
- results/baseline-dropout0.05/l2r_n1_eval/summary.jsonl +26 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-1152.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-1440.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-1728.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2016.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2304.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2592.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-288.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2880.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-3168.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-3456.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-3744.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4032.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4320.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4608.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4896.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-5184.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-5472.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-576.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-5760.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-6048.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-6336.json +32 -0
- results/baseline-dropout0.10/l2r_n1_eval/checkpoint-6624.json +32 -0
.gitattributes
CHANGED
|
@@ -1300,3 +1300,12 @@ runs/t1a/t1a_S2_C3_rand5_r2l50_i5exp/checkpoint-6552/tokenizer.json filter=lfs d
|
|
| 1300 |
runs/t1a/t1a_S2_C2_r2l50_i5exp/checkpoint-3240/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1301 |
runs/t1a/t1a_S2_C1_rand15/checkpoint-2160/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1302 |
runs/t1a/t1a_S2_C0_baseline/checkpoint-1224/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1300 |
runs/t1a/t1a_S2_C2_r2l50_i5exp/checkpoint-3240/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1301 |
runs/t1a/t1a_S2_C1_rand15/checkpoint-2160/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1302 |
runs/t1a/t1a_S2_C0_baseline/checkpoint-1224/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1303 |
+
runs/baseline-dropout0.05/checkpoint-2592/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1304 |
+
runs/baseline-dropout0.05/checkpoint-5472/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1305 |
+
runs/baseline-dropout0.05/checkpoint-2880/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1306 |
+
runs/baseline-dropout0.05/checkpoint-7200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1307 |
+
runs/baseline-dropout0.05/checkpoint-5760/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1308 |
+
runs/baseline-dropout0.05/checkpoint-4896/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1309 |
+
runs/baseline-dropout0.05/checkpoint-5184/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1310 |
+
runs/baseline-dropout0.05/checkpoint-3456/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 1311 |
+
runs/baseline-dropout0.05/checkpoint-1440/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-1152.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-1152",
|
| 4 |
+
"checkpoint": "checkpoint-1152",
|
| 5 |
+
"global_step": 1152,
|
| 6 |
+
"eval_loss": 3.9354641437530518,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9354641437530518,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5061,
|
| 11 |
+
"eval_samples_per_second": 108.46,
|
| 12 |
+
"eval_steps_per_second": 6.798,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-1440.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-1440",
|
| 4 |
+
"checkpoint": "checkpoint-1440",
|
| 5 |
+
"global_step": 1440,
|
| 6 |
+
"eval_loss": 3.876405715942383,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.876405715942383,
|
| 9 |
+
"eval_model_preparation_time": 0.002,
|
| 10 |
+
"eval_runtime": 22.5151,
|
| 11 |
+
"eval_samples_per_second": 108.416,
|
| 12 |
+
"eval_steps_per_second": 6.795,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-1728.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-1728",
|
| 4 |
+
"checkpoint": "checkpoint-1728",
|
| 5 |
+
"global_step": 1728,
|
| 6 |
+
"eval_loss": 3.8658761978149414,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.8658761978149414,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.555,
|
| 11 |
+
"eval_samples_per_second": 108.224,
|
| 12 |
+
"eval_steps_per_second": 6.783,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2016.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2016",
|
| 4 |
+
"checkpoint": "checkpoint-2016",
|
| 5 |
+
"global_step": 2016,
|
| 6 |
+
"eval_loss": 3.865213632583618,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.865213632583618,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5217,
|
| 11 |
+
"eval_samples_per_second": 108.384,
|
| 12 |
+
"eval_steps_per_second": 6.793,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2304.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2304",
|
| 4 |
+
"checkpoint": "checkpoint-2304",
|
| 5 |
+
"global_step": 2304,
|
| 6 |
+
"eval_loss": 3.8903088569641113,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.8903088569641113,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5061,
|
| 11 |
+
"eval_samples_per_second": 108.46,
|
| 12 |
+
"eval_steps_per_second": 6.798,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2592.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2592",
|
| 4 |
+
"checkpoint": "checkpoint-2592",
|
| 5 |
+
"global_step": 2592,
|
| 6 |
+
"eval_loss": 3.9200522899627686,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9200522899627686,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.6157,
|
| 11 |
+
"eval_samples_per_second": 107.934,
|
| 12 |
+
"eval_steps_per_second": 6.765,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-288.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-288",
|
| 4 |
+
"checkpoint": "checkpoint-288",
|
| 5 |
+
"global_step": 288,
|
| 6 |
+
"eval_loss": 5.636253833770752,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 5.636253833770752,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5231,
|
| 11 |
+
"eval_samples_per_second": 108.378,
|
| 12 |
+
"eval_steps_per_second": 6.793,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-2880.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2880",
|
| 4 |
+
"checkpoint": "checkpoint-2880",
|
| 5 |
+
"global_step": 2880,
|
| 6 |
+
"eval_loss": 3.951524019241333,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.951524019241333,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5611,
|
| 11 |
+
"eval_samples_per_second": 108.195,
|
| 12 |
+
"eval_steps_per_second": 6.782,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-3168.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-3168",
|
| 4 |
+
"checkpoint": "checkpoint-3168",
|
| 5 |
+
"global_step": 3168,
|
| 6 |
+
"eval_loss": 3.987661838531494,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.987661838531494,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5622,
|
| 11 |
+
"eval_samples_per_second": 108.19,
|
| 12 |
+
"eval_steps_per_second": 6.781,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-3456.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-3456",
|
| 4 |
+
"checkpoint": "checkpoint-3456",
|
| 5 |
+
"global_step": 3456,
|
| 6 |
+
"eval_loss": 4.019375324249268,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.019375324249268,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.4995,
|
| 11 |
+
"eval_samples_per_second": 108.491,
|
| 12 |
+
"eval_steps_per_second": 6.8,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-3744.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-3744",
|
| 4 |
+
"checkpoint": "checkpoint-3744",
|
| 5 |
+
"global_step": 3744,
|
| 6 |
+
"eval_loss": 4.051281452178955,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.051281452178955,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5569,
|
| 11 |
+
"eval_samples_per_second": 108.215,
|
| 12 |
+
"eval_steps_per_second": 6.783,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4032.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4032",
|
| 4 |
+
"checkpoint": "checkpoint-4032",
|
| 5 |
+
"global_step": 4032,
|
| 6 |
+
"eval_loss": 4.082316875457764,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.082316875457764,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5954,
|
| 11 |
+
"eval_samples_per_second": 108.031,
|
| 12 |
+
"eval_steps_per_second": 6.771,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4320.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4320",
|
| 4 |
+
"checkpoint": "checkpoint-4320",
|
| 5 |
+
"global_step": 4320,
|
| 6 |
+
"eval_loss": 4.112714767456055,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.112714767456055,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.6305,
|
| 11 |
+
"eval_samples_per_second": 107.863,
|
| 12 |
+
"eval_steps_per_second": 6.761,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4608.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4608",
|
| 4 |
+
"checkpoint": "checkpoint-4608",
|
| 5 |
+
"global_step": 4608,
|
| 6 |
+
"eval_loss": 4.137436866760254,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.137436866760254,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.4918,
|
| 11 |
+
"eval_samples_per_second": 108.528,
|
| 12 |
+
"eval_steps_per_second": 6.802,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-4896.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4896",
|
| 4 |
+
"checkpoint": "checkpoint-4896",
|
| 5 |
+
"global_step": 4896,
|
| 6 |
+
"eval_loss": 4.160444259643555,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.160444259643555,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.542,
|
| 11 |
+
"eval_samples_per_second": 108.287,
|
| 12 |
+
"eval_steps_per_second": 6.787,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-5184.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-5184",
|
| 4 |
+
"checkpoint": "checkpoint-5184",
|
| 5 |
+
"global_step": 5184,
|
| 6 |
+
"eval_loss": 4.18454122543335,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.18454122543335,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5997,
|
| 11 |
+
"eval_samples_per_second": 108.01,
|
| 12 |
+
"eval_steps_per_second": 6.77,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-5472.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-5472",
|
| 4 |
+
"checkpoint": "checkpoint-5472",
|
| 5 |
+
"global_step": 5472,
|
| 6 |
+
"eval_loss": 4.198305130004883,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.198305130004883,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5485,
|
| 11 |
+
"eval_samples_per_second": 108.255,
|
| 12 |
+
"eval_steps_per_second": 6.785,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-576.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-576",
|
| 4 |
+
"checkpoint": "checkpoint-576",
|
| 5 |
+
"global_step": 576,
|
| 6 |
+
"eval_loss": 4.593662738800049,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.593662738800049,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.461,
|
| 11 |
+
"eval_samples_per_second": 108.677,
|
| 12 |
+
"eval_steps_per_second": 6.812,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-5760.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-5760",
|
| 4 |
+
"checkpoint": "checkpoint-5760",
|
| 5 |
+
"global_step": 5760,
|
| 6 |
+
"eval_loss": 4.224238395690918,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.224238395690918,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.6107,
|
| 11 |
+
"eval_samples_per_second": 107.958,
|
| 12 |
+
"eval_steps_per_second": 6.767,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6048.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6048",
|
| 4 |
+
"checkpoint": "checkpoint-6048",
|
| 5 |
+
"global_step": 6048,
|
| 6 |
+
"eval_loss": 4.242297649383545,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.242297649383545,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5354,
|
| 11 |
+
"eval_samples_per_second": 108.319,
|
| 12 |
+
"eval_steps_per_second": 6.789,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6336.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6336",
|
| 4 |
+
"checkpoint": "checkpoint-6336",
|
| 5 |
+
"global_step": 6336,
|
| 6 |
+
"eval_loss": 4.2556233406066895,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.2556233406066895,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5517,
|
| 11 |
+
"eval_samples_per_second": 108.24,
|
| 12 |
+
"eval_steps_per_second": 6.784,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6624.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6624",
|
| 4 |
+
"checkpoint": "checkpoint-6624",
|
| 5 |
+
"global_step": 6624,
|
| 6 |
+
"eval_loss": 4.266598224639893,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.266598224639893,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5903,
|
| 11 |
+
"eval_samples_per_second": 108.055,
|
| 12 |
+
"eval_steps_per_second": 6.773,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-6912.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6912",
|
| 4 |
+
"checkpoint": "checkpoint-6912",
|
| 5 |
+
"global_step": 6912,
|
| 6 |
+
"eval_loss": 4.286227703094482,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.286227703094482,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5354,
|
| 11 |
+
"eval_samples_per_second": 108.318,
|
| 12 |
+
"eval_steps_per_second": 6.789,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-72.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-72",
|
| 4 |
+
"checkpoint": "checkpoint-72",
|
| 5 |
+
"global_step": 72,
|
| 6 |
+
"eval_loss": 7.489798069000244,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 7.489798069000244,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 23.9233,
|
| 11 |
+
"eval_samples_per_second": 102.034,
|
| 12 |
+
"eval_steps_per_second": 6.395,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-7200.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-7200",
|
| 4 |
+
"checkpoint": "checkpoint-7200",
|
| 5 |
+
"global_step": 7200,
|
| 6 |
+
"eval_loss": 4.306715965270996,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.306715965270996,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.614,
|
| 11 |
+
"eval_samples_per_second": 107.942,
|
| 12 |
+
"eval_steps_per_second": 6.766,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/checkpoint-864.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-864",
|
| 4 |
+
"checkpoint": "checkpoint-864",
|
| 5 |
+
"global_step": 864,
|
| 6 |
+
"eval_loss": 4.100483417510986,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.100483417510986,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5187,
|
| 11 |
+
"eval_samples_per_second": 108.399,
|
| 12 |
+
"eval_steps_per_second": 6.794,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.05/l2r_n1_eval/summary.jsonl
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-72", "checkpoint": "checkpoint-72", "global_step": 72, "eval_loss": 7.489798069000244, "metrics": {"eval_loss": 7.489798069000244, "eval_model_preparation_time": 0.0018, "eval_runtime": 23.9233, "eval_samples_per_second": 102.034, "eval_steps_per_second": 6.395, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 2 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-288", "checkpoint": "checkpoint-288", "global_step": 288, "eval_loss": 5.636253833770752, "metrics": {"eval_loss": 5.636253833770752, "eval_model_preparation_time": 0.0018, "eval_runtime": 22.5231, "eval_samples_per_second": 108.378, "eval_steps_per_second": 6.793, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 3 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-576", "checkpoint": "checkpoint-576", "global_step": 576, "eval_loss": 4.593662738800049, "metrics": {"eval_loss": 4.593662738800049, "eval_model_preparation_time": 0.0018, "eval_runtime": 22.461, "eval_samples_per_second": 108.677, "eval_steps_per_second": 6.812, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 4 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-864", "checkpoint": "checkpoint-864", "global_step": 864, "eval_loss": 4.100483417510986, "metrics": {"eval_loss": 4.100483417510986, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5187, "eval_samples_per_second": 108.399, "eval_steps_per_second": 6.794, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 5 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-1152", "checkpoint": "checkpoint-1152", "global_step": 1152, "eval_loss": 3.9354641437530518, "metrics": {"eval_loss": 3.9354641437530518, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5061, "eval_samples_per_second": 108.46, "eval_steps_per_second": 6.798, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 6 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-1440", "checkpoint": "checkpoint-1440", "global_step": 1440, "eval_loss": 3.876405715942383, "metrics": {"eval_loss": 3.876405715942383, "eval_model_preparation_time": 0.002, "eval_runtime": 22.5151, "eval_samples_per_second": 108.416, "eval_steps_per_second": 6.795, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 7 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-1728", "checkpoint": "checkpoint-1728", "global_step": 1728, "eval_loss": 3.8658761978149414, "metrics": {"eval_loss": 3.8658761978149414, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.555, "eval_samples_per_second": 108.224, "eval_steps_per_second": 6.783, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 8 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2016", "checkpoint": "checkpoint-2016", "global_step": 2016, "eval_loss": 3.865213632583618, "metrics": {"eval_loss": 3.865213632583618, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5217, "eval_samples_per_second": 108.384, "eval_steps_per_second": 6.793, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 9 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2304", "checkpoint": "checkpoint-2304", "global_step": 2304, "eval_loss": 3.8903088569641113, "metrics": {"eval_loss": 3.8903088569641113, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5061, "eval_samples_per_second": 108.46, "eval_steps_per_second": 6.798, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 10 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2592", "checkpoint": "checkpoint-2592", "global_step": 2592, "eval_loss": 3.9200522899627686, "metrics": {"eval_loss": 3.9200522899627686, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.6157, "eval_samples_per_second": 107.934, "eval_steps_per_second": 6.765, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 11 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-2880", "checkpoint": "checkpoint-2880", "global_step": 2880, "eval_loss": 3.951524019241333, "metrics": {"eval_loss": 3.951524019241333, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5611, "eval_samples_per_second": 108.195, "eval_steps_per_second": 6.782, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 12 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-3168", "checkpoint": "checkpoint-3168", "global_step": 3168, "eval_loss": 3.987661838531494, "metrics": {"eval_loss": 3.987661838531494, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5622, "eval_samples_per_second": 108.19, "eval_steps_per_second": 6.781, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 13 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-3456", "checkpoint": "checkpoint-3456", "global_step": 3456, "eval_loss": 4.019375324249268, "metrics": {"eval_loss": 4.019375324249268, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.4995, "eval_samples_per_second": 108.491, "eval_steps_per_second": 6.8, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 14 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-3744", "checkpoint": "checkpoint-3744", "global_step": 3744, "eval_loss": 4.051281452178955, "metrics": {"eval_loss": 4.051281452178955, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5569, "eval_samples_per_second": 108.215, "eval_steps_per_second": 6.783, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 15 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4032", "checkpoint": "checkpoint-4032", "global_step": 4032, "eval_loss": 4.082316875457764, "metrics": {"eval_loss": 4.082316875457764, "eval_model_preparation_time": 0.0018, "eval_runtime": 22.5954, "eval_samples_per_second": 108.031, "eval_steps_per_second": 6.771, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 16 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4320", "checkpoint": "checkpoint-4320", "global_step": 4320, "eval_loss": 4.112714767456055, "metrics": {"eval_loss": 4.112714767456055, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.6305, "eval_samples_per_second": 107.863, "eval_steps_per_second": 6.761, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 17 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4608", "checkpoint": "checkpoint-4608", "global_step": 4608, "eval_loss": 4.137436866760254, "metrics": {"eval_loss": 4.137436866760254, "eval_model_preparation_time": 0.0018, "eval_runtime": 22.4918, "eval_samples_per_second": 108.528, "eval_steps_per_second": 6.802, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 18 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-4896", "checkpoint": "checkpoint-4896", "global_step": 4896, "eval_loss": 4.160444259643555, "metrics": {"eval_loss": 4.160444259643555, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.542, "eval_samples_per_second": 108.287, "eval_steps_per_second": 6.787, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 19 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-5184", "checkpoint": "checkpoint-5184", "global_step": 5184, "eval_loss": 4.18454122543335, "metrics": {"eval_loss": 4.18454122543335, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5997, "eval_samples_per_second": 108.01, "eval_steps_per_second": 6.77, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 20 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-5472", "checkpoint": "checkpoint-5472", "global_step": 5472, "eval_loss": 4.198305130004883, "metrics": {"eval_loss": 4.198305130004883, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5485, "eval_samples_per_second": 108.255, "eval_steps_per_second": 6.785, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 21 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-5760", "checkpoint": "checkpoint-5760", "global_step": 5760, "eval_loss": 4.224238395690918, "metrics": {"eval_loss": 4.224238395690918, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.6107, "eval_samples_per_second": 107.958, "eval_steps_per_second": 6.767, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 22 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6048", "checkpoint": "checkpoint-6048", "global_step": 6048, "eval_loss": 4.242297649383545, "metrics": {"eval_loss": 4.242297649383545, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5354, "eval_samples_per_second": 108.319, "eval_steps_per_second": 6.789, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 23 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6336", "checkpoint": "checkpoint-6336", "global_step": 6336, "eval_loss": 4.2556233406066895, "metrics": {"eval_loss": 4.2556233406066895, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5517, "eval_samples_per_second": 108.24, "eval_steps_per_second": 6.784, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 24 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6624", "checkpoint": "checkpoint-6624", "global_step": 6624, "eval_loss": 4.266598224639893, "metrics": {"eval_loss": 4.266598224639893, "eval_model_preparation_time": 0.0018, "eval_runtime": 22.5903, "eval_samples_per_second": 108.055, "eval_steps_per_second": 6.773, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 25 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-6912", "checkpoint": "checkpoint-6912", "global_step": 6912, "eval_loss": 4.286227703094482, "metrics": {"eval_loss": 4.286227703094482, "eval_model_preparation_time": 0.0019, "eval_runtime": 22.5354, "eval_samples_per_second": 108.318, "eval_steps_per_second": 6.789, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
| 26 |
+
{"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05", "checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.05/checkpoint-7200", "checkpoint": "checkpoint-7200", "global_step": 7200, "eval_loss": 4.306715965270996, "metrics": {"eval_loss": 4.306715965270996, "eval_model_preparation_time": 0.0018, "eval_runtime": 22.614, "eval_samples_per_second": 107.942, "eval_steps_per_second": 6.766, "epoch": 0.0}, "objective": "LM loss", "add_l2r_token": false, "l2r_percent": 100.0, "logical_per_device_eval_batch_size": 256, "per_device_eval_micro_batch_size": 8, "eval_gradient_accumulation_steps": 32, "world_size_env": 2, "world_size_runtime": 2, "global_logical_eval_batch": 512, "global_eval_batch_size_arg": 512, "batch_size_per_device_arg": 64, "dataloader_num_workers": 4, "dataloader_prefetch_factor": 4, "dataloader_pin_memory": true, "remove_unused_columns": true, "next_i_enabled": false, "max_next_i_train": 1}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-1152.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-1152",
|
| 4 |
+
"checkpoint": "checkpoint-1152",
|
| 5 |
+
"global_step": 1152,
|
| 6 |
+
"eval_loss": 3.9190073013305664,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9190073013305664,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.594,
|
| 11 |
+
"eval_samples_per_second": 108.038,
|
| 12 |
+
"eval_steps_per_second": 6.772,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-1440.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-1440",
|
| 4 |
+
"checkpoint": "checkpoint-1440",
|
| 5 |
+
"global_step": 1440,
|
| 6 |
+
"eval_loss": 3.840430974960327,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.840430974960327,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.4804,
|
| 11 |
+
"eval_samples_per_second": 108.583,
|
| 12 |
+
"eval_steps_per_second": 6.806,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-1728.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-1728",
|
| 4 |
+
"checkpoint": "checkpoint-1728",
|
| 5 |
+
"global_step": 1728,
|
| 6 |
+
"eval_loss": 3.809433937072754,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.809433937072754,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5492,
|
| 11 |
+
"eval_samples_per_second": 108.252,
|
| 12 |
+
"eval_steps_per_second": 6.785,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2016.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-2016",
|
| 4 |
+
"checkpoint": "checkpoint-2016",
|
| 5 |
+
"global_step": 2016,
|
| 6 |
+
"eval_loss": 3.806412696838379,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.806412696838379,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5869,
|
| 11 |
+
"eval_samples_per_second": 108.072,
|
| 12 |
+
"eval_steps_per_second": 6.774,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2304.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-2304",
|
| 4 |
+
"checkpoint": "checkpoint-2304",
|
| 5 |
+
"global_step": 2304,
|
| 6 |
+
"eval_loss": 3.8121891021728516,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.8121891021728516,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5588,
|
| 11 |
+
"eval_samples_per_second": 108.206,
|
| 12 |
+
"eval_steps_per_second": 6.782,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2592.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-2592",
|
| 4 |
+
"checkpoint": "checkpoint-2592",
|
| 5 |
+
"global_step": 2592,
|
| 6 |
+
"eval_loss": 3.8249197006225586,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.8249197006225586,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5514,
|
| 11 |
+
"eval_samples_per_second": 108.242,
|
| 12 |
+
"eval_steps_per_second": 6.785,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-288.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-288",
|
| 4 |
+
"checkpoint": "checkpoint-288",
|
| 5 |
+
"global_step": 288,
|
| 6 |
+
"eval_loss": 5.595135688781738,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 5.595135688781738,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5092,
|
| 11 |
+
"eval_samples_per_second": 108.445,
|
| 12 |
+
"eval_steps_per_second": 6.797,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-2880.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-2880",
|
| 4 |
+
"checkpoint": "checkpoint-2880",
|
| 5 |
+
"global_step": 2880,
|
| 6 |
+
"eval_loss": 3.8407106399536133,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.8407106399536133,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5802,
|
| 11 |
+
"eval_samples_per_second": 108.104,
|
| 12 |
+
"eval_steps_per_second": 6.776,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-3168.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-3168",
|
| 4 |
+
"checkpoint": "checkpoint-3168",
|
| 5 |
+
"global_step": 3168,
|
| 6 |
+
"eval_loss": 3.863334894180298,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.863334894180298,
|
| 9 |
+
"eval_model_preparation_time": 0.002,
|
| 10 |
+
"eval_runtime": 22.5555,
|
| 11 |
+
"eval_samples_per_second": 108.222,
|
| 12 |
+
"eval_steps_per_second": 6.783,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-3456.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-3456",
|
| 4 |
+
"checkpoint": "checkpoint-3456",
|
| 5 |
+
"global_step": 3456,
|
| 6 |
+
"eval_loss": 3.879568338394165,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.879568338394165,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5503,
|
| 11 |
+
"eval_samples_per_second": 108.247,
|
| 12 |
+
"eval_steps_per_second": 6.785,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-3744.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-3744",
|
| 4 |
+
"checkpoint": "checkpoint-3744",
|
| 5 |
+
"global_step": 3744,
|
| 6 |
+
"eval_loss": 3.901733160018921,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.901733160018921,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5126,
|
| 11 |
+
"eval_samples_per_second": 108.428,
|
| 12 |
+
"eval_steps_per_second": 6.796,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4032.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-4032",
|
| 4 |
+
"checkpoint": "checkpoint-4032",
|
| 5 |
+
"global_step": 4032,
|
| 6 |
+
"eval_loss": 3.9239118099212646,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9239118099212646,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5782,
|
| 11 |
+
"eval_samples_per_second": 108.113,
|
| 12 |
+
"eval_steps_per_second": 6.776,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4320.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-4320",
|
| 4 |
+
"checkpoint": "checkpoint-4320",
|
| 5 |
+
"global_step": 4320,
|
| 6 |
+
"eval_loss": 3.9365298748016357,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9365298748016357,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5678,
|
| 11 |
+
"eval_samples_per_second": 108.163,
|
| 12 |
+
"eval_steps_per_second": 6.78,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4608.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-4608",
|
| 4 |
+
"checkpoint": "checkpoint-4608",
|
| 5 |
+
"global_step": 4608,
|
| 6 |
+
"eval_loss": 3.9618711471557617,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.9618711471557617,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5129,
|
| 11 |
+
"eval_samples_per_second": 108.427,
|
| 12 |
+
"eval_steps_per_second": 6.796,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-4896.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-4896",
|
| 4 |
+
"checkpoint": "checkpoint-4896",
|
| 5 |
+
"global_step": 4896,
|
| 6 |
+
"eval_loss": 3.974947452545166,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.974947452545166,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.6206,
|
| 11 |
+
"eval_samples_per_second": 107.911,
|
| 12 |
+
"eval_steps_per_second": 6.764,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-5184.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-5184",
|
| 4 |
+
"checkpoint": "checkpoint-5184",
|
| 5 |
+
"global_step": 5184,
|
| 6 |
+
"eval_loss": 3.994370222091675,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 3.994370222091675,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5457,
|
| 11 |
+
"eval_samples_per_second": 108.269,
|
| 12 |
+
"eval_steps_per_second": 6.786,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-5472.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-5472",
|
| 4 |
+
"checkpoint": "checkpoint-5472",
|
| 5 |
+
"global_step": 5472,
|
| 6 |
+
"eval_loss": 4.0052008628845215,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.0052008628845215,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5653,
|
| 11 |
+
"eval_samples_per_second": 108.175,
|
| 12 |
+
"eval_steps_per_second": 6.78,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-576.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-576",
|
| 4 |
+
"checkpoint": "checkpoint-576",
|
| 5 |
+
"global_step": 576,
|
| 6 |
+
"eval_loss": 4.540627956390381,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.540627956390381,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.5447,
|
| 11 |
+
"eval_samples_per_second": 108.274,
|
| 12 |
+
"eval_steps_per_second": 6.787,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-5760.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-5760",
|
| 4 |
+
"checkpoint": "checkpoint-5760",
|
| 5 |
+
"global_step": 5760,
|
| 6 |
+
"eval_loss": 4.017905235290527,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.017905235290527,
|
| 9 |
+
"eval_model_preparation_time": 0.0018,
|
| 10 |
+
"eval_runtime": 22.554,
|
| 11 |
+
"eval_samples_per_second": 108.229,
|
| 12 |
+
"eval_steps_per_second": 6.784,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-6048.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-6048",
|
| 4 |
+
"checkpoint": "checkpoint-6048",
|
| 5 |
+
"global_step": 6048,
|
| 6 |
+
"eval_loss": 4.0377888679504395,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.0377888679504395,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5986,
|
| 11 |
+
"eval_samples_per_second": 108.016,
|
| 12 |
+
"eval_steps_per_second": 6.77,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-6336.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-6336",
|
| 4 |
+
"checkpoint": "checkpoint-6336",
|
| 5 |
+
"global_step": 6336,
|
| 6 |
+
"eval_loss": 4.044261455535889,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.044261455535889,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.5384,
|
| 11 |
+
"eval_samples_per_second": 108.304,
|
| 12 |
+
"eval_steps_per_second": 6.788,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|
results/baseline-dropout0.10/l2r_n1_eval/checkpoint-6624.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10",
|
| 3 |
+
"checkpoint_path": "/home/ubuntu/workspace/data-augmentations-for-pretraining/runs/baseline-dropout0.10/checkpoint-6624",
|
| 4 |
+
"checkpoint": "checkpoint-6624",
|
| 5 |
+
"global_step": 6624,
|
| 6 |
+
"eval_loss": 4.070178031921387,
|
| 7 |
+
"metrics": {
|
| 8 |
+
"eval_loss": 4.070178031921387,
|
| 9 |
+
"eval_model_preparation_time": 0.0019,
|
| 10 |
+
"eval_runtime": 22.6005,
|
| 11 |
+
"eval_samples_per_second": 108.007,
|
| 12 |
+
"eval_steps_per_second": 6.77,
|
| 13 |
+
"epoch": 0.0
|
| 14 |
+
},
|
| 15 |
+
"objective": "LM loss",
|
| 16 |
+
"add_l2r_token": false,
|
| 17 |
+
"l2r_percent": 100.0,
|
| 18 |
+
"logical_per_device_eval_batch_size": 256,
|
| 19 |
+
"per_device_eval_micro_batch_size": 8,
|
| 20 |
+
"eval_gradient_accumulation_steps": 32,
|
| 21 |
+
"world_size_env": 2,
|
| 22 |
+
"world_size_runtime": 2,
|
| 23 |
+
"global_logical_eval_batch": 512,
|
| 24 |
+
"global_eval_batch_size_arg": 512,
|
| 25 |
+
"batch_size_per_device_arg": 64,
|
| 26 |
+
"dataloader_num_workers": 4,
|
| 27 |
+
"dataloader_prefetch_factor": 4,
|
| 28 |
+
"dataloader_pin_memory": true,
|
| 29 |
+
"remove_unused_columns": true,
|
| 30 |
+
"next_i_enabled": false,
|
| 31 |
+
"max_next_i_train": 1
|
| 32 |
+
}
|